Skip to content

Latest commit

 

History

History
154 lines (111 loc) · 5.12 KB

README.md

File metadata and controls

154 lines (111 loc) · 5.12 KB
Raw

Step-by-Step

This document describes the step-by-step instructions for reproducing PyTorch ResNet50/ResNet18/ResNet101 tuning results with Intel® Low Precision Optimization Tool.

Note

PyTorch quantization implementation in imperative path has limitation on automatically execution. It requires to manually add QuantStub and DequantStub for quantizable ops, it also requires to manually do fusion operation. Intel® Low Precision Optimization Tool has no capability to solve this framework limitation. Intel® Low Precision Optimization Tool supposes user have done these two steps before invoking Intel® Low Precision Optimization Tool interface. For details, please refer to https://pytorch.org/docs/stable/quantization.html

Prerequisite

1. Installation

pip install -r requirements.txt

2. Prepare Dataset

Download ImageNet Raw image to dir: /path/to/imagenet.

Run

1. ResNet50

cd examples/pytorch/image_recognition/imagenet
python main.py -t -a resnet50 --pretrained /path/to/imagenet

2. ResNet18

cd examples/pytorch/image_recognition/imagenet
python main.py -t -a resnet18 --pretrained /path/to/imagenet

3. ResNext101_32x8d

cd examples/pytorch/image_recognition/imagenet
python main.py -t -a resnext101_32x8d --pretrained /path/to/imagenet

4. InceptionV3

cd examples/pytorch/image_recognition/imagenet
python main.py -t -a inception_v3 --pretrained /path/to/imagenet

5. Mobilenet_v2

cd examples/pytorch/image_recognition/imagenet
python main.py -t -a mobilenet_v2 --pretrained /path/to/imagenet

Examples of enabling Intel® Low Precision Optimization Tool auto tuning on PyTorch ResNet

This is a tutorial of how to enable a PyTorch classification model with Intel® Low Precision Optimization Tool.

User Code Analysis

Intel® Low Precision Optimization Tool supports three usages:

  1. User only provide fp32 "model", and configure calibration dataset, evaluation dataset and metric in model-specific yaml config file.
  2. User provide fp32 "model", calibration dataset "q_dataloader" and evaluation dataset "eval_dataloader", and configure metric in tuning.metric field of model-specific yaml config file.
  3. User specifies fp32 "model", calibration dataset "q_dataloader" and a custom "eval_func" which encapsulates the evaluation dataset and metric by itself.

As ResNet18/50/101 series are typical classification models, use Top-K as metric which is built-in supported by Intel® Low Precision Optimization Tool. So here we integrate PyTorch ResNet with Intel® Low Precision Optimization Tool by the first use case for simplicity.

Write Yaml Config File

In examples directory, there is a template.yaml. We could remove most of items and only keep mandotory item for tuning.

framework:
  - name: pytorch                              # possible values are tensorflow, mxnet and pytorch

tuning:
  metric:
    topk: 1                                    # tuning metrics: accuracy 
  accuracy_criterion:
    - relative: 0.01                           # the tuning target of accuracy loss percentage: 1%
  timeout: 0                                   # tuning timeout (seconds)
  random_seed: 9527                            # random seed

calibration:
    iterations: 10
    dataloader:
      batch_size: 30
      dataset:
        - type: "ImageFolder"
        - root: "../imagenet/img/train" # NOTICE: config to your imagenet data path
      transform:
        RandomResizedCrop:
          - size: 224
        RandomHorizontalFlip:
        ToTensor:
        Normalize:
          - mean: [0.485, 0.456, 0.406]
          - std: [0.229, 0.224, 0.225]

evaluation:
  dataloader:
    batch_size: 30
    dataset:
      - type: "ImageFolder"
      - root: "../imagenet/img/val" # NOTICE: config to your imagenet data path
    transform:
      Resize:
        - size: 256
      CenterCrop:
        - size: 224
      ToTensor:
      Normalize:
        - mean: [0.485, 0.456, 0.406]
        - std: [0.229, 0.224, 0.225]

Here we choose topk built-in metric and set accuracy target as tolerating 0.01 relative accuracy loss of baseline. The default tuning strategy is basic strategy. The timeout 0 means unlimited time for a tuning config meet accuracy target.

Prepare

PyTorch quantization requires two manual steps:

  1. Add QuantStub and DeQuantStub for all quantizable ops.
  2. Fuse possible patterns, such as Conv + Relu and Conv + BN + Relu.

Torchvision provide quantized_model, so we didn't do these steps above for all torchvision models. Please refer torchvision

The related code please refer to examples/pytorch/image_recognition/resnet/main.py.

Code Update

After prepare step is done, we just need update main.py like below.

model.eval()
model.module.fuse_model()
import ilit
tuner = ilit.Tuner("./conf.yaml")
q_model = tuner.tune(model)

The tune() function will return a best quantized model during timeout constrain.